There are 740,000 annual events of syncope (transient loss of consciousness) that lead to an emergency department (ED) visit, resulting in 250,000 admissions and $2.4 billion in yearly hospital costs. There are currently no effective prediction tools to identify older patients (agee60 years) who may be at risk for subsequent cardiac death or serious cardiac events. As a result, up to 85% of older adults with syncope are hospitalized for diagnostic evaluation. However, there is no evidence that admission improves diagnostic yield, quality-of-life, or mortality. Fundamental problems with the existing paradigm of care include inaccurate risk assessment and the lack of an evidence base to match predicted risk with appropriate clinical actions. Furthermore, existing research is virtually silent on how t risk stratify and evaluate older adults, who disproportionately account for diagnostic uncertainty, health service use, and serious outcomes. As a result, practice patterns have not changed over the past 30 years. Mounting pressures to constrain health care costs increase the topicality of this problem, and syncope was recently identified as one of the top conditions targeted by Medicare Recovery Audit Contractors for repossession of medically unnecessary inpatient expenditures. Innovative care models must be developed to improve the efficiency and value of the ED evaluation. To address these needs, we propose a prospective, observational study of 3,700 older adults with unexplained syncope enrolled from four emergency departments. Our specific hypotheses are that: 1.) current patterns of care are costly with low clinical benefit; 2. explicit criteria will improve risk stratification compared to unstructured physician assessment and published risk models; and 3.) risk-based decision-making can safely reduce costs compared to existing care. To assess these hypotheses, our proposal has the following sequential Aims: Specific Aim 1. Describe rates, diagnostic yield, therapeutic yield, and costs of diagnostic admission and testing associated with existing care. Specific Aim 2. Derive and validate a novel risk prediction model for 30-day cardiac death and serious cardiac outcomes after unexplained syncope. Specific Aim 3. Estimate diagnostic yield and costs of implementing risk-based decision algorithms. Our study will result in innovative care algorithms for a common and costly syndrome, and completion of our Aims will facilitate a paradigm shift in the evaluation of syncope in older adults.